What is to blame for America’s segregated schools? Housing patterns and school attendance boundaries play big roles, certainly. But some have also pointed the finger at another culprit: charter schools.

An Associated Press analysis from 2017 found that about one in six charter schools were severely segregated, and implied that the schools were deepening the isolation of students of color. Presidential candidate Bernie Sanders cited that report in his education plan, concluding, “The damage to communities caused by unregulated charter school growth must be stopped and reversed.”

Charter advocates have pushed back, arguing that charter schools aren’t the cause of that segregation — they have only tried to provide better options for students in already segregated areas.

Now, a first-of-its-kind national study offers some of the clearest answers to date about whether charter schools are contributing to racial segregation.

The researchers find that the expansion of charters has in fact made segregation worse within school districts. But the impact is quite small: if charters were to vanish tomorrow, segregation in school districts across the country would fall by 5 to 7%.

“Ninety-five percent of the segregation is still there,” said Brian Kisida of the University of Missouri, who conducted the research with Tomas Monarrez and Matt Chingos of the Urban Institute.

Looking across entire metropolitan areas, which usually include many school districts, charter schools didn’t clearly increase or decrease racial segregation.

“Our study shows that critics are incorrect when they say that charters are driving a resegregation of American schools,” wrote the three researchers. “But it also shows that charter proponents are incorrect to assume that freeing public schools from neighborhood boundaries will necessarily enhance racial integration.”

Researchers go beyond simple comparisons to try to nail down the effect of charters

Nationally, about a third of charter school students are Hispanic, a third are white, and a quarter are black, but those students aren’t evenly spread across schools, as the Associated Press and others have pointed out.

It’s quite challenging to track how charter schools actually affect school segregation, though. Simply looking at whether charters are more or less segregated than traditional district schools at one point in time, as the AP did, doesn’t show whether school systems are made more segregated because of charters.

To get around this, the latest study looked at the racial makeup of nearly every school in the country between 1998 and 2015. Then researchers compared trends in racial segregation at different grade levels in each school district. The logic is that, if charter schools were increasing racial segregation, you would expect segregation to increase more in the grade levels with more charters. (The study does not consider economic segregation, which has increased nationally in recent years.)

The paper concludes that, yes, charters really do make racial segregation worse, but only modestly. The researchers measure segregation on a scale of 0 (a student’s race isn’t predictive at all of the race of his or her school peers) to 100 (a student’s race is entirely predictive).

“If the average district in the sample shut down all of its charter schools, we would expect its overall segregation of black and Hispanic students to decline from 15.0 to 14.2,” the researchers explain — a decrease of 5 percent.

One reason this effect is so small is that in most districts, few students attend charters. Looking only at districts with charter schools, eliminating charters would cut segregation by 7 percent.

The effect differs from state to state. In some, charters made things worse, and in others, there was no clear effect. That’s similar to a number of past state-level studies.

“We find substantial variation, with evidence of notable effects on segregation in Louisiana, New Mexico, North Carolina, Oklahoma, and Rhode Island,” the researchers write. “And there are several states where charters appear to have little or no effect on segregation, such as Arizona, Florida, Georgia, New Jersey, and Oregon.”

So far, we’ve only been talking about segregation within individual districts. But segregation is generally greater between districts in the same area — think a city and its surrounding suburbs. When the researchers expand their scope to a metropolitan area, the segregation effect of charters largely disappears. Charters seem to modestly decrease segregation between school districts, particularly in areas where metro areas are fragmented into many districts.

That may reflect more white families living within less-white school districts but sending their children to predominantly white charter schools. It also echoes other research showing that untethering schools from neighborhoods through school choice, including charters, encourages more integrated housing but not schools. (The number of predominantly white charters nationwide remains relatively low.)

For black students in particular, though, charter schools continue to exacerbate segregation when looking across a metro area, but the effect is again quite small.

“Taken together, we find compelling evidence that the rise of charter schools over the last 20 years has led to slightly higher levels of racial and ethnic segregation, on average,” the researchers conclude.

The academic paper was released by the Urban Institute, a D.C.-based think tank, and an abridged version was published by Education Next, a Harvard-based journal that is generally sympathetic to charter schools. (The study was funded by the Walton Family Foundation, which backs charter schools. Chalkbeat is also supported by Walton.)

External researchers who reviewed the study at Chalkbeat’s request said they found the results compelling.

“This is the first credible national study of the effects of charter schools on racial integration,” said Brian Gill, a researcher at Mathematica who has extensively studied charter schools. “This is a methodologically strong study that the field really needed.”

Ann Owens, a researcher on segregation at the University of Southern California, said the results mirror a paper she has in progress herself.

The findings are important in part because American schools remain sharply segregated by race even as research has consistently linked more integrated schools to better academic outcomes for black and Hispanic students. Monarrez, Kisida and Chingos note that themselves, but argue that shouldn’t be the end of the discussion.

“A large number of charter schools were founded and specifically tailored to serve students from vulnerable backgrounds, out of which a good number have been successful at improving student outcomes,” they write.

Research has found that charter schools perform about the same as nearby district schools, but often do better in cities and for students of color, including networks like KIPP that predominantly serve low-income black and Hispanic students. A small number of “diverse-by-design” charters — 125 schools, according to one estimate, in a sector of several thousand — are actively seeking to create integrated environments.

Penn State University professor Erica Frankenberg argues that being a school of choice doesn’t solve the problem of racial isolation.

“I think that segregation is a concern no matter what type of school it is in,” said Frankenberg, who studies school segregation. “I think it’s one of the things that we should consider when we evaluate charter schools, in addition to a range of social and academic outcomes.”

And Owens of USC says that even the modest increases in segregation caused by charters are troubling — and could grow as charters continue to expand.

“A 5-7% decline in school segregation is not nothing,” she said. “There’s plenty more opportunity for segregation to continue to increase or be sustained by charters.”